کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
407046 | 678124 | 2014 | 12 صفحه PDF | دانلود رایگان |
Selective ensemble is a learning paradigm that follows an “overproduce and choose” strategy, where a number of candidate classifiers are trained, and a set of several classifiers that are accurate and diverse are selected to solve a problem. In this paper, the hybrid approach called D3C is presented; this approach is a hybrid model of ensemble pruning that is based on k-means clustering and the framework of dynamic selection and circulating in combination with a sequential search method. Additionally, a multi-label D3C is derived from D3C through employing a problem transformation for multi-label classification. Empirical study shows that D3C exhibits competitive performance against other high-performance methods, and experiments in multi-label datasets verify the feasibility of multi-label D3C.
Journal: Neurocomputing - Volume 123, 10 January 2014, Pages 424–435